Direct Learning to Rank and Rerank
By Cynthia Rudin, Yining Wang. 2018
11 citations recorded by the source corpus.
Learning-to-rank techniques have proven to be extremely useful for prioritization problems, where we rank items in order of their estimated probabilities, and dedicate our limited resources to the top-ranked items. This work exposes a serious problem with the state of learning-to-rank algorithms, which is that they are based on convex proxies that lead to poor approximations. We then discuss the possibility of "exact" reranking algorithms based on mathematical programming. We prove that a relaxed version of the "exact" problem has the same optimal solution, and provide an empirical analysis.
Subfields: Artificial Intelligence, Computer Networks and Communications, Management Science and Operations Research
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Cynthia Rudin
- Chudi Zhong
- Margo I. Seltzer
- Jimmy J. Lin
- Diane Hu
- Tong Wang
- Zeynep Akkalyoncu Yilmaz
- Lesia Semenova
In Computer Networks and Communications
- Jimmy J. Lin
- Margo I. Seltzer
- Cynthia Rudin
- Allison Chang
- Tianbao Yang
- Robert E. Schapire
- Yiming Ying
- Boris G. Mirkin
In Management Science and Operations Research
Citations in the corpus
- Generalized and Scalable Optimal Sparse Decision Trees (2020, methodology): “Optimizing a convex proxy for a rank statistic can yield results that are far from optimal (see Rudin & Wang, 2018).”